Device Classification Service Using Expiration Timers

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current device classification methods in computer networks, particularly in IoT environments, face challenges in accurately classifying devices due to subtle behavioral differences, leading to misclassifications and the need for immediate policy application, while long-term observation is impractical for real-time security and access control.

Innovation Solution

A device classification service that applies a first network policy with an expiration timer, allowing for progressive refinement of device classifications based on ongoing telemetry data analysis, enabling reclassification and adjustment of policies without sandboxing, thereby improving accuracy and handling re-classifications incrementally.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of time

If device classification is based on short-term behavioral observation (e.g., first minute), then policy application can be immediate, but classification accuracy deteriorates due to subtle behavioral differences

Engineering Contradiction:
Improvetime to apply policyVSAvoidclassification accuracy
Core Design Contradiction:
Loss of timeVSMeasurement precision

Solution Approach 1:

The system performs preliminary classification based on limited initial data (first minute observation) to enable immediate policy application, while simultaneously preparing for refinement as more data becomes available. This allows the system to act promptly without waiting for complete information.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The classification is made dynamic through expiration timers and progressive refinement. The classification validity period is set to expire, allowing the system to re-evaluate and update classifications as more behavioral data accumulates, transitioning from static short-term classification to dynamic long-term refinement.

Inventive Principle:
Principle #15Dynamics

2Measurement precision

If long-term behavioral observation is used for classification, then classification accuracy improves, but the ability to apply appropriate policies at all points in the process deteriorates

Engineering Contradiction:
Improveclassification accuracyVSAvoidtime to apply policy
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The system applies preliminary policies based on initial classification before long-term observation is complete. These policies are applied with expiration timers that allow them to be superseded by refined classifications later, ensuring policies are available immediately but can be improved over time.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system continuously monitors device behavior and uses feedback from ongoing observation to refine initial classifications. When classifications are updated based on accumulated data, the system can adjust policies accordingly, creating a closed-loop system that improves accuracy over time while maintaining immediate policy enforcement.

Inventive Principle:
Principle #23Feedback

3Measurement precision

If device classification is refined progressively using colored device trees, then classification accuracy improves, but device complexity increases

Engineering Contradiction:
Improveclassification accuracyVSAvoidclassification system complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The classification system is segmented into hierarchical levels using colored device trees, where each color represents a different classification confidence level or refinement stage. This segmentation allows the system to manage complexity by breaking down the classification process into manageable, color-coded categories that can be applied incrementally.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system adds a temporal dimension to classification by introducing expiration timers and progressive refinement stages. Classifications are no longer static binary states but evolve over time through multiple refinement levels, adding a time-based dimension that manages complexity through structured progression rather than simultaneous complex evaluation.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

Data Source

PatentEP3745671B1Progressive refinement of device classifications using colored device and policy trees
Publication Date: 2022.09.14 CISCO TECHNOLOGY INC
  • EP3745671B1 patent drawingFigure 1A
  • EP3745671B1 patent drawingFigure 1B
  • EP3745671B1 patent drawingFigure 2

AI summary

In embodiments, a device classification service classifies a device in a network as being of a first device type. The service applies a first network policy that has an associated expiration timer to the device, based on its classification as being of the first device type. The service determines whether the device was reclassified as being of a different device type than that of the first device type before expiration of the expiration timer associated with the first network policy. The service applies a second network policy to the device, when the service determines that the device has not been reclassified as being of a different device type before expiration of the expiration timer associated with the first network policy.